Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2009-05-01

00:00:02webguest60nvm i found it
00:01:03webguest60It works!!!!!! thank you so much
00:01:19webguest60I can now watch Red vs Blue in French class
00:03:33_fmlI have r20830, nothing modified
00:04:17_fmlOn target, r20812, the scrolling works
00:04:33webguest60what player?
00:04:54sharkoshi folks; is the "root.m3u8" playlist a standard part of any rockbox distribution? because I can't find it with a little fgrep, so where does it come from?
00:05:08Llorean_fml: Are the scrolling settings the same?
00:05:15Lloreansharkos: It shows up when you make it.
00:05:32 Quit EternalRains1 (Read error: 104 (Connection reset by peer))
00:05:48sharkosLlorean: thanks; so is it a default option in the text field or something?
00:06:06Lloreansharkos: No, there's an option to create a playlist that recursively scans all subfolders and names the playlist after the folder you're in
00:06:07BigBambiIt is what the playlist is named by default if you make one of everything on the player
00:06:10LloreanIf you're in root, it names it root.m3u8
00:06:27LloreanWell, recursively scans if that's enabled
00:06:33LloreanOtherwise it just uses the files in that folder, I think
00:06:49 Join fyrestorm [0] (
00:07:05webguest60Do you guys get paid for helping people all day?
00:07:17 Quit EternalRains (Read error: 110 (Connection timed out))
00:07:24BigBambiRockbox is all 100% volunteer - development, support, etc.
00:07:29_fmlLlorean: I think yes. It's the auto scrolling in the file browser (not in WPS) which should be activated automatically. Are there some settings? The only maybe related commit since 20812 is r20829. I'll try to compile r20828.
00:07:36webguest60Oh yeah good point...
00:07:55webguest60mabey get some adds on the site for some cash.....
00:08:33scorchewebguest60: there is always that little paypal donation button ;)
00:09:32webguest60I will never give information like that over the internet....
00:09:46webguest60It's not safe
00:10:17BigBambiIt is safer than over the phone/in a restaurant/etc., but anyway, that is off-topic
00:10:56webguest60Well just one more thing to add to that, how is it safer then in a restraurant also i don't buy over the phone either
00:11:13BigBambino, it is off-topic
00:11:25scorchewebguest60: if you want to talk about this further, type "/join #rockbox-community" ;)(
00:12:13webguest60are you in that chat also?
00:12:23webguest60or only one at a time?
00:12:29BigBambialso there
00:12:33webguest60ok good
00:13:24_fmlr80829 seems to be the offender. With r20828, the lines scroll, with r20829 not anymore.
00:14:58_fmlkugel: are reading this?
00:15:06_fml*are you
00:19:20_fmlLlorean: what do you think: should I file a bug report?
00:20:45LloreanIf something's broken, file it. :)
00:21:04 Quit jgarvey ("Leaving")
00:21:28 Nick MT is now known as mt (n=MTee@
00:22:06 Quit bertrik (Read error: 113 (No route to host))
00:28:23 Quit itcheg (" ajax IRC Client")
00:29:25 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
00:32:35 Join itcheg [0] (i=62cb408a@gateway/web/ajax/
00:38:34***Saving seen data "./dancer.seen"
00:39:52 Quit moredhel (Client Quit)
00:40:08 Join faemir [0] (
00:42:27 Quit itcheg (" ajax IRC Client")
00:44:12 Quit _fml ("CGI:IRC 0.5.9 (2006/06/06)")
00:54:25 Quit tchan (Read error: 60 (Operation timed out))
00:57:53 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
00:59:37 Quit moos ("Rockbox rules the DAP world")
01:01:34 Join robin0800 [0] (
01:03:20 Quit linuxstb (Read error: 113 (No route to host))
01:03:33 Quit webguest60 ("CGI:IRC (EOF)")
01:04:51 Quit robin0800 (Client Quit)
01:05:32 Join robin0800 [0] (
01:06:43 Quit ender (" It always takes longer than you expect, even when you take Hofstadter's Law into account. -- Hofstadter's Law")
01:07:04 Join itcheg [0] (i=62cb408a@gateway/web/ajax/
01:11:21 Quit robin0800 ("No Ping reply in 90 seconds.")
01:12:01 Join robin0800 [0] (
01:13:32 Quit bubsy (Read error: 54 (Connection reset by peer))
01:16:11 Quit Lss (Read error: 60 (Operation timed out))
01:17:36 Quit domonoky (Read error: 104 (Connection reset by peer))
01:19:01 Quit Thundercloud (Remote closed the connection)
01:19:14 Join fdinel [0] (
01:20:40 Join bubsy [0] (i=Bubsy@
01:22:25 Quit perrikwp (" ajax IRC Client")
01:22:36 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
01:23:37 Quit itcheg (" ajax IRC Client")
01:25:23 Join itcheg [0] (i=62cb408a@gateway/web/ajax/
01:29:05 Quit JdGordon| (" ajax IRC Client")
01:30:29 Join cmwslw [0] (
01:49:44 Quit itcheg (" ajax IRC Client")
01:50:16 Join EternalRains [0] (
01:53:07 Quit flux (Read error: 104 (Connection reset by peer))
01:53:14 Join flux [0] (
01:53:33 Quit faemir (
01:53:33 Quit preglow (
01:53:33 Quit scorche|sh (
01:53:33 Quit bittin```` (
01:53:34 Quit crculver (
01:53:34 Quit daurnimator (
01:53:34 Quit rvvs89 (
01:53:34 Quit kkurbjun (
01:53:34 Quit Tuplanolla (
01:53:34 Quit GodEater (
01:53:34 Quit Kopfgeldjaeger (
01:53:34 Quit yosafbridge (
01:53:34 Quit jhulst (
01:53:35 Quit suom1 (
01:53:35 Quit fyrestorm (
01:53:35 Quit freqmod (
01:53:35 Quit markun (
01:53:35 Quit Galois (
01:53:35 Quit Beta2K (
01:53:35 Quit Bombe (
01:53:35 Quit fxb__ (
01:53:35 Quit liiwi (
01:54:07NJoinfaemir [0] (
01:54:07NJoinpreglow [0] (i=thomj@rockbox/developer/preglow)
01:54:07NJoinbittin```` [0] (
01:54:07NJoinscorche|sh [50] (n=scorche@rockbox/administrator/scorche)
01:54:46NJoinfyrestorm [0] (
01:54:46NJoinfreqmod [0] (
01:54:46NJoinmarkun [50] (n=markun@rockbox/developer/markun)
01:54:46NJoinGalois [0] (
01:54:46NJoinBeta2K [0] (
01:54:46NJoinsuom1 [0] (
01:54:46NJoinBombe [0] (n=droden@freenet/developer/Bombe)
01:54:46NJoinfxb__ [0] (
01:54:46NJoinliiwi [0] (
01:54:53NJoincrculver [0] (
01:54:53NJoindaurnimator [0] (n=daurnima@unaffiliated/daurnimator)
01:54:53 Join rvvs89 [0] (n=ivo@pdpc/supporter/base/rvvs89)
01:54:53NJoinkkurbjun [0] (n=kkurbjun@rockbox/developer/kkurbjun)
01:54:53NJoinTuplanolla [0] (n=jani@unaffiliated/tuplanolla)
01:54:53NJoinGodEater [0] (n=nnyeahri@rockbox/staff/GodEater)
01:54:53NJoinKopfgeldjaeger [0] (
01:54:53NJoinyosafbridge [0] (
01:54:53 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
01:57:29 Quit sharkos ("Leaving.")
01:59:18 Quit amiconn (Nick collision from services.)
01:59:20 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
01:59:40 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
02:00:08 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
02:00:08 Quit pixelma (Nick collision from services.)
02:00:28 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
02:12:22 Join Tristan [0] (
02:14:56 Join evilnick_bs [0] (i=620ec27e@gateway/web/ajax/
02:35:55 Quit kachna (Read error: 113 (No route to host))
02:38:38***Saving seen data "./dancer.seen"
02:38:57 Join CaptainKwel [0] (
02:40:46 Quit robin0800 ("No Ping reply in 90 seconds.")
02:41:26 Join robin0800 [0] (
02:52:49 Quit evilnick_bs (" ajax IRC Client")
03:04:39 Join itcheg [0] (i=62db4767@gateway/web/ajax/
03:14:18 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
03:29:04 Quit robin0800 (Read error: 60 (Operation timed out))
03:55:08 Join daurn [0] (n=daurnima@unaffiliated/daurnimator)
04:00:30 Quit daurnimator (Read error: 60 (Operation timed out))
04:10:29 Quit EternalRains (Read error: 104 (Connection reset by peer))
04:12:58Unhelpfulso, "official test builds" would be appropriate for jpeg album art?
04:14:26cmwslwDoes anyone here have an ipod nano second or third generation?
04:14:58scorcheif you have a further question, it is typically best to ask that right out...
04:15:41saratogaUnhelpful: I think so
04:17:28 Quit Zoxc ()
04:20:04LloreanUnhelpful: Yes.
04:20:45cmwslwI need people to post some DFU info about their 2G and 3G nanos
04:21:19cmwslw2G nano's have a DFU-like mode
04:22:06froggymanyeah they do, its on the wiki... somewhere
04:22:22froggymani remeber that you do have destroy the OF though
04:27:46 Quit froggyman ("CGI:IRC")
04:30:48 Quit efyx_ (Remote closed the connection)
04:35:12Unhelpfuli'm *assuming* the main objection about FS #10177 will be size... i'll try to find a few places to reduce that tonight.
04:36:03LloreanUnhelpful: I think the feeling is that the added convenience of supporting (or being closer to supporting) the 'standard' way of doing Album Art outweighs the size issue mostly
04:36:11LloreanOf course, reducing size is always a plus
04:37:35 Quit cmwslw ("Ex-Chat")
04:38:25saratogado we not support folder.jpg?
04:38:41***Saving seen data "./dancer.seen"
04:39:02Unhelpfulsaratoga: i could always add another test for folder.jpg - i assume that should only be for the in-same-folder search?
04:39:05saratogawell cover.jpg wprls great!
04:39:45Unhelpfulgood to know it wrpls for you ;)
04:39:46saratogaUnhelpful: randomly looking at my albums, apparently some software likes to use folder.jpg since I've got a few of those, though maybe its not worth adding to rockbox
04:40:05 Join hd [0] (
04:40:13Unhelpfulisn't folder.jpg some windows convention for providing folder icons?
04:40:34saratogait could be that wmp uses it for album art or something
04:41:14 Quit HellDragon (Nick collision from services.)
04:41:20 Nick hd is now known as HellDragon (n=jd@Wikipedia/HellDragon)
04:45:39saratogaUnhelpful: folder.jpg is the WMP convention, which is probably why i have so much album art using it, we should probably add it to rockbox once your patch goes in (i guess folder.bmp was pointless)
04:46:02 Join robin0800 [0] (
04:46:36Unhelpfula separate test for folder.jpg in the same folder as the file, then?
04:47:05saratogait would be nice
04:47:35 Quit miepchen^schla (Read error: 101 (Network is unreachable))
04:48:57Unhelpfulit's pretty easy to add... if we're only adding it to handle WMP cover images, i don't think it needs to be another option for all of the other cover search paths, and we probably needn't bother looking for folder.bmp, either
04:49:47saratogawould it be possible to look for any jpg file as a last resort in a folder?
04:51:33Unhelpfulit would be rather inconvenient. the other methods are all based on generating some base filename from the path and tags. supporting any jpeg in the same folder as AA would mean reading the entire directory contents, and would probably not be able to guarantee that the same one is found each time for the same directory.
04:52:19saratogaok then never mind, its probably not worth it anyway
04:52:34JdGordonwhy wouldnt it have the same file each time?
04:52:47JdGordonas long as the dirfilter and sort order dont change.. it shoudlnt
04:52:53JdGordonunless of course the files change
04:53:21UnhelpfulJdGordon: does readdir always return sorted results?
04:56:28JdGordonft_load() i think does
04:56:47 Join dreamlayers [0] (n=dreamlay@rockbox/developer/dreamlayers)
04:56:48JdGordonthere are 2 folder reading functions in rockbox... one shouldnt be used expect by the other
05:00:54*Llorean thinks just supporting folder.jpg as a nod to WMP convention is probably good enough
05:01:40LloreanDoes it seem likely people will have absolutely arbitrary image names though?
05:03:03LloreanThat's not arbitrary.
05:03:07Lloreanthat's the one we already support
05:05:29 Join motionman95 [0] (i=4597cd8a@gateway/web/ajax/
05:10:29motionman95Hello? Any rockbox devs online?
05:10:56*JdGordon counts at least 4
05:11:21*motionman95 only sees 2.
05:11:32*motionman95 prepares to ask his question...
05:11:38JdGordonmotionman95: please get into the habbit of just asking... if someone is online they will answer
05:11:43JdGordonand if not they might read the logs
05:14:41motionman95Okay. Anyway, my problem is that my plugin's acting up. It's designed to parse and out data from cfg files. It uses the "read_line" function. The only problem is that when it reads .cfg files, it only outputs the first two letters of the first two lines. I've gone over the code, like 100+ times, and I still can't spot where I went wrong.
05:15:04motionman95Maybe it's because I'm to inexperienced. Should I post my code?
05:15:17JdGordonthat would help...
05:15:21motionman95In the PasteBin? Or upload the .C file?
05:15:39 Quit itcheg (" ajax IRC Client")
05:15:45motionman95PasteBin it is...2 minutes, please!
05:17:56Unhelpfulhm... my e200 takes 2.57s to open a 2560x1600 jpeg. that's not half bad... much better than it'd do scaling a bmp of that size to screen dimensions.
05:18:22CIA-63New commit by dreamlayers (r20834): Fix another FS #10130 issue - Set backlight and button light brightness via the backlight thread to avoid potential concurrency issues.
05:18:55 Join mib_fvw7qy [0] (i=4597cd8a@gateway/web/ajax/
05:19:00LloreanUnhelpful: Does it use any intelligence to pick a good nearby size to open, then scale to the desired size, or does it just open fullsize then scale down?
05:20:18UnhelpfulLlorean: it selects the smallest scalefactor which will produce output the same as, or larger than, the desired size. that's a *huge* part of the savings, since 1/8 scaling is much, much cheaper than properly decoding the image.
05:21:27saratogathe 1/8 DCT is just the average value right?
05:21:44LloreanUnhelpful: That's more or less what I was asking if it did. :)
05:21:49motionman95@JdGordon: There's the link...^
05:22:09Unhelpfulsaratoga: right - you add a rounding factor to the DC coefficient and shift right, clamp to range, and done.
05:23:08 Quit l403 (Read error: 113 (No route to host))
05:24:13motionman95 ....
05:24:31LloreanYou don't need to repeat it
05:24:48*motionman95 agoligizes..
05:25:02*motionman95 corrects himself...
05:25:03Unhelpfulthe 2/8 scaling is *nearly* as simple, since there are no scaling factors to multiply by
05:25:52Unhelpfuli think, given how much space it takes up now, that it's not a great idea to add the non-power-of-two scaling factors
05:27:50 Quit robin0800 ("No Ping reply in 90 seconds.")
05:28:29 Join robin0800 [0] (
05:28:36motionman95Is anyone willing to help me?... I think JdGordan left...
05:29:06dreamlayersmotionman95: I'll take a look at it have the link?
05:34:04dreamlayersmotionman95: Yes I have the link
05:37:33 Quit gevaerts (Nick collision from services.)
05:37:45 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
05:38:10dreamlayersmotionman95: remove_whitespace has a problem: it doesn't copy the last character or put a terminating null on noSpaces. As for rb->read_line, it behaves properly; it reads lines as expected.
05:38:46*motionman95 check his code...
05:39:26JdGordonmotionman95: sorry, was on the phone
05:40:26motionman95@dreamlayers: If I tell it to copy the last character, do I still need to null terminate it?
05:40:57motionman95@JdGordon: That's okay...
05:41:24dreamlayersmotionman95: You can make it copy the null, by not subtracting 1 from the result of strlen and using i <= len in the for loop.
05:41:49motionman95I'm gonna log out here, and log in from my other laptop. One sec...
05:41:53 Quit motionman95 (" ajax IRC Client")
05:42:27 Part mib_fvw7qy
05:42:41 Join motionman95 [0] (i=4597cd8a@gateway/web/ajax/
05:43:57motionman95Thanks dreamlayers. I'll see if it works now, did you notice anything else?
05:45:29dreamlayersWhy do you subtract 1 from the result of strlen in other functions? Do you not want to look at the last character?
05:46:42dreamlayersIn parseSetting, should be value[i2] = '\0';
05:47:40motionman95Wow. How come I didn't notice any of those mistakes?
05:48:48dreamlayersIt can be easier to see someone else's bugs.
05:51:44motionman95\me tests his plugin...
05:51:55*motionman95 tests his plugin.
05:53:24 Join kachna [0] (
05:53:24*motionman95 cries, his plugin still doesn't work... =(
05:53:56dreamlayersHow is the plugin misbehaving now?
05:55:19motionman95It's doing what it did before. All it's sending to OUTPUT.txt is "/.rockbox/font" when there's other settings it's supposed to parse... =(
05:56:19 Quit robin0800 ("No Ping reply in 90 seconds.")
05:56:33JdGordonmotionman95: hang on.. what are you trying to do? exactly....
05:56:58 Join robin0800 [0] (
05:57:40 Quit yosafbridge ("Coyote finally caught me")
05:58:00 Join yosafbridge [0] (
05:58:19JdGordonyou probaly just want to use settings_parseline() from misc.c which might need to be added to the plugin lib
05:58:36JdGordonno, its there already
06:00:19 Join ark_ [0] (i=ark@
06:02:29ark_hi, I wanted to know about the support for sansa c200 series with firmware V2
06:02:45ark_is it not supported yet?
06:02:55scorcheis it on the front page yet?
06:03:20ark_i'm a little confused about the versions
06:04:05ark_not sure what is my player firmware version
06:04:20scorcherecent development should be posted in the corresponding thread in the forums...also, it is important to note that it isnt just the "V2 formware", but more a massive hardware change...due to this, we try to use the term "AMS sansas" to denote these devices and emphasize this difference
06:05:45ark_ahh, I didn't know that
06:05:58scorcheerrrr....firmware...not formware =)
06:06:17ark_I'm unlucky, I thought I could install rockbox, jaja, yeah, I read firmware
06:06:39ark_I bought thinking about rockbox, but didn't know about the version
06:06:56scorchestay tuned though!....a port is being worked on =)
06:07:14*motionman95 is back...
06:07:16ark_oh, cool, thanks, I'll be
06:13:20 Quit intrados (Read error: 104 (Connection reset by peer))
06:14:30 Join intrados [0] (
06:18:50 Join SirFunk_ [0] (
06:20:32 Join fizekul [0] (n=18079edf@gateway/web/cgi-irc/
06:21:50*motionman95 is tired...will try again tomorrow...
06:21:54 Quit motionman95 (" ajax IRC Client")
06:23:07 Quit dreamlayers ("Goodnight")
06:23:28 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
06:31:53 Quit SirFunk__ (Read error: 110 (Connection timed out))
06:32:12fizekulis anyone here?
06:32:27saratogajust ask a question, the channel is logged
06:32:36fizekulo ok, thanks
06:33:34fizekuli'm trying to install rockbox on my gigabeat s30 from windows and the instruction PDF starts off with me needing to download beastpatcher.exe, which i can't find
06:34:59saratogadid you check the wiki? i'd guess theres a copy on there
06:35:16 Quit SirFunk_ (Read error: 104 (Connection reset by peer))
06:35:50 Join SirFunk_ [0] (
06:36:05Lloreanfizekul: It's not a supported player yet, so it *may* be necessary for you to compile some of the tools yourself.
06:37:07fizekulit says on the wiki "see utils/MTP/beastpatcher in SVN for the latest code"
06:37:38fizekuli checked there and there was only source code, no executable, so i'm guessing this is one of those tools i have to compile on my own
06:37:43saratogathe wiki says we haven't gotten beastpatcher for windows yet, so thats probably why you couldn't find it
06:38:38fizekulso, if i wanted to go about the install, i should try linux?
06:38:44***Saving seen data "./dancer.seen"
06:39:03saratogai think theres some other windows install tool on the wiki
06:41:25fizekulthanks for the help, i'm going to have to spend some time on this later and figure out what i need after i finish up some other work
06:41:31 Quit fizekul ("CGI:IRC")
06:43:20 Join evilnick1 [0] (
06:48:56 Quit SirFunk_ (Read error: 60 (Operation timed out))
06:49:27 Join SirFunk_ [0] (
06:57:12 Quit ark_ (Remote closed the connection)
07:00:10 Quit CaptainKwel (Remote closed the connection)
07:00:34 Quit evilnick (Read error: 113 (No route to host))
07:03:04 Quit robin0800 (Read error: 104 (Connection reset by peer))
07:08:20 Quit SirFunk_ (Read error: 110 (Connection timed out))
07:08:41 Join SirFunk_ [0] (
07:13:31 Join Mastery [0] (
07:18:18 Join SirFunk__ [0] (
07:27:01 Join Lss [0] (
07:27:16 Join SirFunk [0] (
07:30:24 Join Horschti [0] (n=Horscht@xbmc/user/horscht)
07:30:55 Quit Lss (Client Quit)
07:31:16 Join Lss [0] (
07:31:19 Quit SirFunk_ (Read error: 110 (Connection timed out))
07:35:47 Quit jmillikin (Read error: 60 (Operation timed out))
07:41:51 Quit SirFunk__ (Read error: 110 (Connection timed out))
07:48:06 Quit Horscht (Read error: 110 (Connection timed out))
08:06:02 Join bertrik [0] (
08:21:20Unhelpfulis there a reason we only use -O, rather than -Os on ARM? i tried replacing a function pointer with a switch statement (and inlining the called functions, since they're only called there). it costs 240B on ARM with -O, but saves 320B with -Os...
08:21:54LloreanPerformance hits, I think.
08:22:09LloreanI seem to recall someone tested the performances of them.
08:23:08Unhelpfulhave we changed compiler versions since then?
08:23:15 Quit SirFunk (Read error: 110 (Connection timed out))
08:23:32LloreanI think with ARM we've only gone from 4.0.1 to 4.0.2 to 4.0.3
08:23:47LloreanI don't know at which point they were tested.
08:24:06LloreanIn fact, I'm not sure they were tested for anything but codecs, for that matter, and that's only if my memory isn't playing tricks on me.
08:24:19 Join SirFunk [0] (
08:26:07 Join __lifeless [0] (n=lifeless@
08:30:02 Quit BHSPitMonkey (Remote closed the connection)
08:31:09Unhelpfulhm. ok... i don't see many really good space savings left. jpeg_load.o is down to 19KiB on ARM color, 13.8KiB on coldfire color.
08:31:27 Join Rob2222 [0] (
08:37:03 Quit faemir ("Leaving")
08:38:46***Saving seen data "./dancer.seen"
08:40:02 Quit _lifeless (Read error: 110 (Connection timed out))
08:47:26 Quit Zarggg_ (Read error: 110 (Connection timed out))
08:49:05 Quit Rob2223 (Read error: 110 (Connection timed out))
08:55:28 Join flydutch [0] (
08:55:44LloreanGodEater: I don't really get your objection to having the SVN table.
08:55:53LloreanYou've said you don't think it's harmful, but you seem rather vehemently against it.
08:57:31*scorche runs up against FS #9391
09:02:43 Quit Foxx (Read error: 110 (Connection timed out))
09:05:37mtNow no mallocs in cook decoder (it only used one in the init function and it's been replaced with a fixed size array) .. linuxstb : and now the decoder consumes packets directly from the scrambling unit rather than memcpy-ing an input buffer.
09:07:54 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
09:12:43amiconnUnhelpful: Did you add greyscale target support as well (didn't check the fs task, just asking because you only mentioned the colour target binsize increase)?
09:13:36 Quit z35 ("Leaving")
09:13:36Unhelpfulamiconn: yes, greyscale targets are supported, with dequant and idct skipped for the chroma components.
09:16:06 Join Thundercloud [0] (
09:16:59Unhelpfulthe *largest* space improvement, from un-inlining reads the read-in portion of check_bit_buffer, applies to all targets. several smaller space optimizations did not. the size difference on color vs greyscale targets is not all that large.
09:18:07*amiconn is also curious how code size and performance compare to the jpeg plugin
09:20:55 Join Zarggg [0] (
09:20:57Unhelpfulthat's rather difficult to say, just directly comparing objects won't work, because the jpeg plugin also includes all of that slideshow and caching functionality
09:24:18 Nick J-23 is now known as Pierwszejer (
09:24:22 Nick Pierwszejer is now known as J-23 (
09:25:14Unhelpfulthere are 2x as many function-calls-via-pointer as in the plugin, because of the split idct. the idct functions are a bit smaller, because dequant is done as coefficients are unpacked. performance is probably hurt a bit by the output always being passed through the scaler, although the linear scaler at 1:1 is pretty fast
09:29:14amiconnFractional scaling is something I've always wanted in the jpeg viewer
09:31:20Unhelpfuli think this could reasonably be adapted as an in-memory decoder, by using some #defines to substitute the getc/putc implementations for ones that just push a pointer back and forth. they could probably be inlined pretty cheaply, then, too.
09:32:34 Quit linuxstb ("Leaving")
09:32:48 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
09:35:46linuxstbmt: Sounds good! Have you tested the decoder very much? e.g. on lots of different rm files?
09:39:43mtlinuxstb : not after the last couple of modifications .. I just did them 30 minutes ago :)
09:40:16mtI just tested it on one sample, but will test the other 5 now
09:41:19mtI also want to modify the makefile to make it compile and link, since it only does the linking part currently
09:42:43 Quit martian67 (
09:42:43 Quit bzed (
09:42:43 Quit happosade (
09:42:43 Quit dionoea (
09:42:46mtalso to change the ($ffmpeg-config −−cflags) passed to gcc with the actual flags .. ffmpeg installation shouldn't be a requirement for compilation :)
09:43:27NJoinmartian67 [0] (n=martian6@about/linux/regular/martian67)
09:43:27NJoinbzed [0] (
09:43:27NJoinhapposade [0] (
09:43:27NJoindionoea [0] (n=dionoea@videolan/developer/dionoea)
09:43:33 Quit Xerion (Read error: 110 (Connection timed out))
09:43:59markunmt: in the wiki I see you had to go back to an older version of cook.c for the fixed-point patch. Were you able to update it to the latest version ofter that?
09:46:07mtmarkun : Haven't tried yet .. although the latest patches were mostly preparation for multichannel cook, and the current patched version I'm using isn't very different from the more recent one I was using before.
09:49:57mtlinuxstb : just finished testing on all the 7 samples I have, all OK.
09:55:19 Quit Thundercloud (Remote closed the connection)
10:02:49 Quit crculver ("Leaving")
10:04:40linuxstbmt: Did you see my comments to you in this channel a couple of days ago? You left the channel just after I said them, so maybe you missed them...
10:04:52linuxstb(about preparing a patch to be committed to Rockbox SVN)
10:05:26mtno I didn't .. I'll go check the logs
10:10:04mtlinuxstb : ok .. currently I have main() inside cook.c, I'll just have to take that into a separate main.c.
10:11:36mtHowever, The decoder is a bit different from ffmpeg's current one, since it's an old patched revision.
10:20:56 Join l403 [0] (n=l@
10:21:22 Join codesquid [0] (n=quassel@kong.kawo1.RWTH-Aachen.DE)
10:21:48linuxstbmt: What I meant is that it would be nice to commit your current code in various stages, starting with something as close to ffmpeg's original code (the svn revision you took), and then a series of patches to bring it to the current state. But maybe others would think that's a waste of time...
10:22:30mtlinuxstb : ah ok, no problem with that.
10:24:06 Quit kachna (Read error: 113 (No route to host))
10:24:49mtlinuxstb : I always have to chmod a=rwx outputfile.wav - How do I modify the permissions in open() to avoid this ?
10:28:26 Quit __lifeless (Read error: 110 (Connection timed out))
10:34:13linuxstbmt: My flac test program (in Rockbox SVN) uses fd=open(filename,O_CREAT|O_WRONLY|O_TRUNC,S_IRUSR|S_IWUSR);
10:35:47mtS_IRWXU == u=rwx , right ?
10:36:47mtoh yes .. fixed :)
10:38:50***Saving seen data "./dancer.seen"
10:40:24amiconnWhy would a .wav file need the x bit set?
10:43:04 Join Foxx [0] (
10:43:51 Join {phoenix} [0] (
10:44:13 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
10:48:57mtamiconn : it doesn't I was just asking ..
10:57:44Unhelpfulok, things kind of got away from me here. i'll let the jpeg patches sit until tonight, and try to finish up these last couple of optimizations and commit them then, unless i hear more about it by then.
11:00:17 Quit {phoenix} (Remote closed the connection)
11:06:06 Join daurn| [0] (
11:11:22 Join pbxy [0] (n=pbxy@2001:5c0:1000:a:0:0:0:211)
11:13:51 Nick J-23 is now known as pecet (
11:13:54 Nick pecet is now known as J-23 (
11:14:48amiconnUnhelpful: Does jpeg support also work for PF?
11:14:49BigBambiUnhelpful: folder.jpg would be great - as it is something of a windows standard many album art downloading programs default to saving folder.jpg too.
11:16:30 Join mirak [0] (
11:18:08UnhelpfulBigBambi: we'd talked about that before, and i'll probably add a separate test for <dir>/folder.jpg to AA search
11:18:13amiconnLlorean: Did you find your archos usb cable meanwhile?
11:18:56 Join ender` [0] (
11:18:57 Quit pbxy ("Leaving")
11:19:00BigBambiUnhelpful: Cool - I currently have for instance cover.bmp and folder.jpg in each of mine, and it'd be good to dump the duplication :)
11:19:02 Quit daurn (Read error: 110 (Connection timed out))
11:19:52Unhelpfulamiconn: it doesn't, yet. i'll add a pluginlib wrapper that takes the same call signature as read_jpeg_file and read_bmp_file, and calls the latter if the filename ends in .bmp (this is what core is doing, so it should work exactly the same.
11:22:51amiconnThere are the mono targets as well, which will need the jpeg loader code in the pluginlib too. And then there are plugin ram constraints on some targets
11:23:27 Join stoffel [0] (
11:25:17*amiconn wonders whether we should ditch bmp aa support once jpeg aa support is in
11:25:21Unhelpfulindeed - PictureFlow is pretty much exactly the plugin buffer on archos, throwing in the jpeg loader will mean using overlay
11:25:27amiconnHow common is bmp aa outside the rockbox world?
11:25:43BigBambiI've not got anything else that uses it
11:26:20Unhelpfuli wouldn't think very. and jpeg makes it much more reasonable to use the very same AA as on your PC, since you can downscale by up to 8x before the scaler, quickly and with good quality.
11:26:21amiconnWe'll still need the bmp loader of course, for wps, icons etc
11:26:46amiconnBut it wouldn't need to be able to scale anymore
11:27:26 Join miepchen^schlaf [0] (
11:27:28Unhelpfulin that case, might as well just revert bmp.c to pre-scaler ;)
11:28:43 Join hakercai [0] (n=3b2aa4a6@gateway/web/cgi-irc/
11:34:10 Quit hakercai ("CGI:IRC (Ping timeout)")
11:35:44 Join pbxy [0] (n=pbxy@2001:5c0:1000:a:0:0:0:185)
11:38:52 Join Grahack [0] (
11:43:34 Join midijunkie [0] (
11:49:42 Quit midijunkie (Read error: 104 (Connection reset by peer))
11:51:16GrahackHi there, I'd like to know if someone wrote a utility to check playlists. Thanks.
11:53:09LloreanGrahack: I don't know what you mean by "check playlists" but since Rockbox uses a standard playlist format (m3u / m3u8) any such utility wouldn't really be Rockbox specific and you're probably better off asking somewhere more focused on music management
11:53:19LloreanThis channel is for Rockbox support and development, not third-party tools
11:54:39Grahackok thanks
11:56:06 Quit fyrestorm (Read error: 104 (Connection reset by peer))
12:03:52 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
12:04:07 Join EternalRains [0] (
12:08:43 Join fyrestorm [0] (
12:09:25Unhelpfulamiconn: re: binsizes, i've not compared rockbox-info.txt, but the loader is a single file, and compiles to 19176B on ARM color, 18864B on ARM greyscale, 14228B on coldfire color, and 13756B on coldfire greyscale
12:13:16 Join _lifeless [0] (n=lifeless@
12:14:48Unhelpfuli can save ~400B of that on ARM, and ~600B of that on coldfire, by moving descaling of the IDCT output out of the IDCT functions. this means another loop that iterates over each of those values, though. it looks like it costs 6-15% extra decode time, depending on scale factors, etc.
12:17:57 Join evilnick [0] (
12:20:50amiconnEfficiency is also important. If the buffers are in dram, extra store/load pairs are especially bad on coldfire
12:22:11 Quit EternalRains (Read error: 104 (Connection reset by peer))
12:23:44Unhelpfulthe image buffer is taken from whatever buffer space the caller provides, but it doesn't get any extra stores with that change. the idct buffer is on stack, and i suspect that the hit is from storing and retrieving the data in it as ints one more time, and in having another loop. i'll leave this change out for commit.
12:36:14 Quit evilnick1 (Read error: 113 (No route to host))
12:38:51***Saving seen data "./dancer.seen"
12:45:35 Join efyx_ [0] (
12:51:56 Quit _lifeless (Remote closed the connection)
12:52:13 Join _lifeless [0] (n=lifeless@
12:56:05 Join barrywardell [0] (
13:00:51 Quit bertrik (Read error: 113 (No route to host))
13:03:30 Quit martian67 (Read error: 110 (Connection timed out))
13:30:36 Join moos [0] (i=mustapha@rockbox/staff/moos)
13:34:48 Join ibseco [0] (
13:58:00 Quit mt (Read error: 104 (Connection reset by peer))
13:58:33 Join mt [0] (n=MTee@
13:58:37 Join __lifeless [0] (n=lifeless@
14:03:35 Quit _lifeless (Read error: 60 (Operation timed out))
14:12:40 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
14:14:39 Join einhirn [0] (
14:15:12 Nick Horschti is now known as Horscht (n=Horscht@xbmc/user/horscht)
14:15:47 Join LambdaCalculus37 [0] (i=44a0430d@rockbox/staff/LambdaCalculus37)
14:24:32 Join ibseco_ [0] (
14:27:03 Quit Grahack ("Leaving.")
14:27:35 Join billiob [0] (
14:28:38billiobhi, does anyone know if a rockbox device (iriver hp120) can be used as an usb bootable device?
14:31:42LambdaCalculus37You could use one as a USB device.
14:32:29billiobok, thx, i'll try it
14:32:40 Join jgarvey [0] (
14:38:54***Saving seen data "./dancer.seen"
14:40:26 Quit ibseco (Read error: 110 (Connection timed out))
14:46:54 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
14:52:24 Quit miepchen^schlaf ()
15:02:13 Quit evilnick_230 (" ajax IRC Client")
15:06:47 Part billiob
15:08:57 Quit einhirn (Read error: 54 (Connection reset by peer))
15:14:10 Join evilnick_7 [0] (i=0c140464@gateway/web/ajax/
15:19:27 Join itcheg [0] (i=41d59de2@gateway/web/ajax/
15:25:37 Join petur [50] (n=petur@rockbox/developer/petur)
15:27:37 Quit efyx_ (Client Quit)
15:29:39 Join miepchen^schlaf [0] (
15:32:30 Join efyx_ [0] (
15:33:55evilnick_7What does "unsupported 0" mean when opening a .jpg file in Rockbox? The file still opens after a pause
15:39:05LambdaCalculus37evilnick_7: Probably the way the JPG was encoded?
15:44:02 Quit SirFunk (Read error: 60 (Operation timed out))
16:06:26 Quit DataGhost (Nick collision from services.)
16:06:34 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
16:12:40 Join motionman95 [0] (i=4597cd8a@gateway/web/ajax/
16:14:24motionman95Okay. My problem is that my plugin's acting up. It's designed to parse and out data from cfg files. It uses the "read_line" function. The only problem is that when it reads .cfg files, it only outputs the first two letters of the first two lines. I've gone over the code, like 100+ times, and I still can't spot where I went wrong.
16:15:07motionman95Repeat of last night's to my plugin source code...
16:16:37 Quit DataGhost (Nick collision from services.)
16:16:45 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
16:20:00 Join faemir [0] (
16:24:29*motionman95 wonders if anyone is online who could help him...
16:26:40 Quit fyrestorm (Read error: 104 (Connection reset by peer))
16:27:42BigBambimotionman95: As with every time you ask a question, just be patient. Do not repeatedly ask, or wonder if people are there, or anything else. Just wait and if someone that can help you sees the question I am sure they will speak up.
16:28:13*motionman95 thinks that sounds like a good idea...
16:28:27 Join fyrestorm [0] (
16:28:50BigBambimotionman95: And you say it sounds look a doog idea everytime someone tells you this. How about this time you actually act on it, and don't just think it is a good idea?
16:29:16BigBambithank you
16:36:02jhulstoops, wrong window, sorry
16:38:55***Saving seen data "./dancer.seen"
16:39:34domonokymotionman95: and where in this code do you output the read text ?
16:45:11domonokymotionman95: also why this static noSpaces buffer? why not just copy it back to the given buffer in remove_whitespace ?
16:48:13domonokyalso "if(cfg_settings.wps_file)" when wps_file is a char array, is probably not what you want...
16:54:20 Quit codesquid (Remote closed the connection)
17:00:05 Join Lear [0] (i=chatzill@rockbox/developer/lear)
17:01:24 Join Zoxc [0] (
17:03:04 Join toffe82 [0] (n=chatzill@
17:05:09 Quit kadoban (Remote closed the connection)
17:07:28 Join kadoban [0] (
17:07:29 Join Xerion [0] (
17:20:13 Join Danceman [0] (n=Danceman@
17:26:44 Quit SUSaiyan` ()
17:34:06 Join BUMBACL0T [0] (n=ORF@unaffiliated/bumbacl0t)
17:34:41 Part BUMBACL0T
17:43:29motionman95domonoky: You still there? What to you mean copy it back to the given buffer?
17:44:57 Quit evilnick_7 (" ajax IRC Client")
17:45:42domonokymotionman95: if you strcpy it back to "string" instead of "nospaces" in remove_whitespace() you can remove the static nospaces.
17:45:47 Join SUSaiyan [0] (n=SUSaiyan@
17:47:36domonokycurrently you copy the whitespace removed string into nospaces in remove_spaces() and copy it back to cur_line in loadSettings, thats not-needed if you copy directly back to string.
17:48:00JdGordonmotionman95: bah, i told oyu last night to just use settings_parseline()
17:48:56domonokybut the real issue in this code, is probably the completly wrong write out in plugin_start..
17:50:17 Quit tvelocity (Remote closed the connection)
17:53:08motionman95JdGordon: I tried using that function, but I can't figure out how it works.
17:54:02saratogaFS #10180 - Better runtime for PP502x target through running at 24MHz
17:54:48 Join evilnick_7 [0] (i=0c140464@gateway/web/ajax/
17:55:01 Quit Danceman (Read error: 110 (Connection timed out))
17:57:30saratogawhat happened to the idea of using zero wait boosting on PP?
18:03:35*LambdaCalculus37 wonders if he should mention FS #10177 and FS #10178 on the AlbumArt wiki page
18:03:59LambdaCalculus37Those two patches are Unhelpful's work in progress to use JPEG album art.
18:04:20BigBambiI wouldn't bother
18:04:29BigBambiThey will go in very soon I think
18:04:42BigBambiAnd then the page can be updated to say jpgs etc.
18:05:18LambdaCalculus37BigBambi: Just remind me to buy Unhelpful a beer for his hard work. ;)
18:12:08 Quit dfkt (Read error: 54 (Connection reset by peer))
18:13:21 Quit ibseco_ (Read error: 145 (Connection timed out))
18:15:42 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
18:17:46 Join ibseco [0] (
18:23:09 Quit Lear ("ChatZilla 0.9.84 [Firefox 3.5b4/20090423204732]")
18:24:11motionman95Question: The keypad on my C250 is acting up. The Select button acts like the down button, for some obscure reason.
18:24:35motionman95I was thinking of maybe buying a broken C250, and using its keypad.
18:28:06 Join stoffel_ [0] (
18:28:34 Quit wincent (Read error: 110 (Connection timed out))
18:29:45motionman95I commented out all the other lines that write to the text file in plugin_start, and only left the line "rb->write(fd, cfg_settings.wps_file, MAX_LEN);". Now the only thing written to the text file is "/". =(
18:30:07 Join tvelocity [0] (
18:36:37 Join midijunkie [0] (
18:38:03 Quit motionman95 (" ajax IRC Client")
18:38:58***Saving seen data "./dancer.seen"
18:39:38 Join jmillikin [0] (n=jmilliki@
18:40:51 Join motionman95 [0] (i=4597cd8a@gateway/web/ajax/
18:40:52 Quit stoffel (Read error: 110 (Connection timed out))
18:41:14evilnick_7wrong window again ')
18:44:20 Quit stoffel_ ("leaving")
18:46:09motionman95How do i build the sim>
18:46:44saratogasame as a target except you choose sim instead of normal in the configure script
18:46:58BigBambiand then after "make" you do "make install"
18:47:28 Join JdGordon| [0] (i=836b0070@gateway/web/ajax/
18:51:33motionman95What does:
18:51:34motionman95 error: `exit' redeclared as different kind of symbol
18:51:54 Quit l403 ("Ex-Chat")
18:51:55motionman95the line is "bool exit = false"
18:52:25saratogai don't think you can have a bool named exit
18:53:19fluxmotionman95, there's already a function called 'exit'
19:01:28 Join fatahab [0] (
19:08:01 Quit __lifeless (Read error: 60 (Operation timed out))
19:08:23 Join _lifeless [0] (n=lifeless@
19:21:12motionman95How do I run "rockboxui"?
19:21:35evilnick_7Is that the sim build?
19:21:45motionman95I hink
19:21:49motionman95I think
19:22:07evilnick_7You should be able to double-click on it (in Windows)
19:22:44motionman95I says: choose the program you want to use to open this file.
19:23:05BigBambidid you cross compile it?
19:23:23BigBambiI'll take that as a no
19:23:30BigBambirockboxui is a linux executable do I cross compile it? I'm using vmware...
19:25:50motionman95Ohh....thanks evilnick_7!
19:29:48motionman95LOL, my plugin doesn't work even in the Sim...
19:33:07 Join __lifeless [0] (n=lifeless@
19:33:22 Quit _lifeless (Remote closed the connection)
19:37:36 Join Grahack [0] (
19:52:03 Join z35 [0] (
19:53:24motionman95I think I'm just gonna delete my whole plguin and start again..
19:54:11 Quit itcheg (" ajax IRC Client")
19:55:53 Join Vorador [0] (
19:57:02 Join Thundercloud [0] (
19:58:00 Quit evilnick_7 (" ajax IRC Client")
20:00:30 Quit Vorador (Client Quit)
20:02:33 Quit pbxy ("Leaving")
20:08:17 Quit barrywardell (Remote closed the connection)
20:10:16GodEaterevening all
20:11:49LambdaCalculus37Evening, GodEater!
20:21:23 Quit midijunkie (Read error: 104 (Connection reset by peer))
20:22:00GrahackHi all and good evening indeed, is there a reason why the generated .m3u8 have a double slash at the beginning of every line ?
20:24:37motionman95Why doesn't rb->splash work in the Sim?
20:25:54 Quit ibseco (Read error: 60 (Operation timed out))
20:36:12GodEaterhi LambdaCalculus37 :)
20:36:28*GodEater hits "Post" on his longest ever forum submission.
20:36:50 Quit miepchen^schlaf ()
20:37:28 Quit flydutch ("/* empty */")
20:38:52*motionman95 reads post...
20:39:00***Saving seen data "./dancer.seen"
20:39:26*Bagder believes in a user-centric front page
20:40:56JdGordon|I can see why it would make sense for a user centric front page... but it should have a dev centric one also
20:41:26Bagderoh indeed
20:41:37Bagderbut I was referring to
20:41:41GodEateryes - I completely agree with having a ""
20:42:04saratogai do not
20:42:16saratogahaving an SVN chart on the front page is sufficient compromise
20:42:28saratogathis is not an unreasonable request, and the arguments against it are stupid
20:42:58BagderI kind of like the idea of letting "users" play with the "user site" view and move over to the "dev parts" and sit there and watch dev details
20:43:27saratogathe problem I see with that is keeping the two groups of people sorted
20:43:43saratogaparticularly when users are simply going to have to click through to the developer site in order to follow development
20:43:59 Join miepchen^schlaf [0] (
20:44:04Bagderyes, I guess the hard part is to please/cover the "middle ground"
20:44:04saratogamore likely what will happen is we will just get more people in the forums asking about things
20:44:54GodEaterand imo an influx of more people who can help them
20:45:27GodEaterof course Llorean's brain will explode as he'll insist on trying to help them all out personally too
20:45:33Bagderalthough, I don't think svn table on the users front or not is a major issue on this
20:45:57saratogaoddly enough, I think its the only real point of disagreement
20:47:03saratogathe new site actually looks quite nice otherwise
20:47:19GodEaterI want it left as either the collapsible feature it is currently, or not there at all.
20:49:23saratogathe argument has been advanced that 1) having an SVN table is somehow intimidating or off putting to users, and 2) we should care about whatever proportion of people who fall under 1)
20:49:46 Quit advcomp2019 (
20:49:46 Quit timc (
20:49:46 Quit Ridayah (
20:49:46 Quit ChanServ (
20:49:46 Quit JdGordon| (
20:49:46 Quit gevaerts (
20:49:46 Quit bagawk (
20:49:46 Quit BlakeJohnson86 (
20:49:46 Quit BigBambi (
20:49:46 Quit scorche (
20:49:46 Quit gromit` (
20:49:47 Quit kadoban (
20:49:47 Quit fyrestorm (
20:49:47 Quit _Auron_ (
20:49:47 Quit Chex (
20:49:47 Quit z35 (
20:49:47 Quit mt (
20:49:47 Quit daurn| (
20:49:47 Quit yosafbridge (
20:49:47 Quit KBH (
20:49:47 Quit planetbeing (
20:49:47 Quit krazykit (
20:49:47 Quit courtc (
20:51:21 Join advcomp2019_ [0] (
20:51:21NJoinz35 [0] (
20:51:21NJoinmt [0] (n=MTee@
20:51:21NJoindaurn| [0] (
20:51:21NJoinyosafbridge [0] (
20:51:21NJoinKBH [0] (
20:51:21NJoinplanetbeing [0] (
20:51:21NJoinkrazykit [0] (
20:51:21NJoincourtc [0] (n=court@unaffiliated/courtc)
20:51:27NJoinkadoban [0] (
20:51:27NJoinfyrestorm [0] (
20:51:27NJoin_Auron_ [0] (
20:51:27NJoinChex [0] (
20:51:34 Quit mt (Success)
20:51:36 Join mt [0] (n=MTee@
20:51:38NJoinJdGordon| [0] (i=836b0070@gateway/web/ajax/
20:51:38NJoingevaerts [0] (n=fg@rockbox/developer/gevaerts)
20:51:38NJoinbagawk [0] (
20:51:38NJoinBlakeJohnson86 [0] (
20:51:38NJoingromit` [0] (
20:51:38NJoinscorche [50] (n=scorche@rockbox/administrator/scorche)
20:51:38NJoinBigBambi [0] (n=alex@rockbox/staff/BigBambi)
20:51:44NJoinRidayah [0] (
20:51:51 Quit motionman95 (" ajax IRC Client")
20:51:51 Quit perrikwp (" ajax IRC Client")
20:52:47NJointimc [0] (n=aoeu@
20:58:32JdGordon|bah, my last message didnt go through according to the logs...
20:58:36JdGordon|how is the svn table relevant to users at all? those that understand it will find the 5 items not enough anyway... and those that dont will just get confused
20:59:00GodEatermy view too
20:59:40gevaertsthe svn table is useful to show people that rockbox is really worked on. If you're used to the usual obsolete sourceforge projects that never actually got started, this is the first thing you look for
20:59:49JdGordon|the *only* reason it might be helpful is to show that the front page isnt comepltly static.. but that doesnt help anyway
21:00:08gevaertswhether or not that means that should be on the front page is a different issue of course
21:00:35 Quit n17ikh ("moving")
21:00:44GodEaterwe can show how a "Rockbox last updated with code at xx:xx" thing on the front page
21:00:49GodEaterit doesn't need to be a commit log
21:00:55BigBambiI'd replace the wiki activity with the svn log
21:01:10gevaertssomething like that could work
21:01:37*domonoky likes the "slide-out" svn table. perhaps it could be named "development" and also contain the link to the dev-page ? (slide-out on hover, goto devpage on click)
21:01:54gevaertsBigBambi: why? I can see the wiki as useful for users, in the sense that they can easily get involved there
21:02:20JdGordon|the wiki is far more interesting to users than the svn table
21:02:28JdGordon|wiki updates*
21:02:46BigBambiI never saw the wiki updates as useful, even when first arriving
21:02:47gevaertsas a non-user I think so anyway :)
21:03:00GodEaterBigBambi: do you have any non-techy friends you could ask ?
21:03:17*gevaerts reads the wiki updates regularly, but he nearly never looks at the svn bit
21:03:32GodEaterindeed, as a dev you have "svn log" at your finger tips
21:03:49JdGordon|I prefer the table
21:03:57gevaertsthat and both -cvs and the bot here
21:04:13BigBambiGodEater: I'll ask a couple on Monday
21:04:34*gevaerts doesn't object to the svn table, but he thinks thatif we get to keep either that or the wiki updates, the wiki updates should win
21:04:44BigBambiBut when I arrived as a new person, I very quickly discovered I cared about cvs, but not the wiki
21:05:02GodEaterbut not right away
21:05:08BigBambiNow it is different - I get svn updates elsewhere and still don't care about the wiki
21:05:18BigBambiGodEater: Right away I didn't know what to look at
21:05:24*toffe82 likes the two tables svn and wiki :)
21:05:24gevaertsAlso, *if* we do a user/dev split, I think the user site shouldn't show -dev mailing list history
21:05:27 Quit intrados (Remote closed the connection)
21:05:37GodEatergevaerts: agreed there too
21:05:55GodEaterif we do a split, we should do it properly
21:14:03NJoinChanServ [0] (ChanServ@services.)
21:14:03Mode"#rockbox +o ChanServ " by
21:14:56 Join motionman95 [0] (i=4597cd8a@gateway/web/ajax/
21:15:41 Quit Zarggg ()
21:15:46motionman95how do you return a char array from a function?
21:17:08GodEaterif the char array is defined in the function you don't
21:19:47JdGordon|well... you can...
21:20:04GodEaternot a good plan though :)#
21:20:24gevaertsGodEater: it could be static
21:20:35 Quit martian67 (Read error: 104 (Connection reset by peer))
21:20:40motionman95What if it was passed through a param?
21:20:51 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
21:22:18saratogadoesn't matter how you pass it
21:23:17 Join Zarggg [0] (
21:35:36LloreanJdGordon|: All the users who've commented in the thread were more interested in SVN updates
21:35:53LloreanSVN updates tell them something about what they *want*, Wiki ones tell them something about what they can *do*. Most users don't come here to do something
21:36:45JdGordon|they are still users.... they dont know what they want... :D
21:37:14Lloreangevaerts: Wiki updates are more interesting to us than users. Users couldn't care less what wiki page was last updated, all they need to know or care about the wiki is if their information is there at all
21:37:18*GodEater remembers at least one user complaining that the svn table didn't contain what he wanted to know
21:37:41*GodEater seems to recall that user made it to the GoldenQuotes page
21:37:51LloreanWhat was it he wanted to know?
21:38:00GodEaterwhy we weren't working on more games :)
21:38:11LloreanI still think that between recent mail, recent wiki, and recent SVN, the SVN is at least the most informative to new people in general
21:39:02LloreanGodEater: Why do you think it shouldn't be there.
21:39:06LloreanYou said you don't think it's harmful
21:39:14LloreanWhat exactly do you have against it, then?
21:39:55GodEater*I* don't - but my poll of people showed it is
21:40:03gevaertsLlorean: I'm not sure if I agree in a regular stable release world
21:40:17GodEaterpeople don't like it
21:40:18Lloreangevaerts: "Oh, something happened two months ago... doesn't seem very active..."
21:40:52LloreanGodEater: Who did you poll besides your girlfriend?
21:41:00GodEaterLlorean: did you read my post ?
21:41:05gevaertsLlorean: yes, that bit of svn is useful. Does it need to be on the front page though, or is a development page that's clearly linked from the front page enough?
21:41:21Lloreangevaerts: The front page is the _project_ front page.
21:41:32GodEaterso it's NOT for users then
21:41:33LloreanIt's what every new person sees. It should immediately show a few things that *everyone* should see
21:41:51LloreanGodEater: Do you have a way of making sure prospective contributors go somewhere else first?
21:41:52*JdGordon| doesnt see the quote GodEater mentinoed
21:42:11*GodEater wonders how firefox et al manage without an svn feed on their front page
21:42:44LloreanGodEater: Firefox has paid contributors.
21:42:49*gevaerts misses something he considers very important on the current front page...
21:43:23gevaertsGodEater: firefox has a huge marketing bulldozer, they don't need an svn feed to convince people that they still exist
21:43:25LloreanGodEater: If we're not dependent upon donations, our biggest asset is new developers. For other projects, users are more of an asset than they are here, changing the balance.
21:43:41GodEaterJdGordon|: ""SVN activity so lame! Its really boring little fixes. When you start work at RockBoy, RockZX, RockNes that all people needed?"
21:44:00GodEaterLlorean: did you read my post yet ?
21:44:32LloreanGodEater: I'm reading it
21:44:42LloreanDoes it answer my question about why we want users more than developers?
21:45:14 Join {phoenix} [0] (
21:45:32LloreanNot really, just finished it.
21:45:36 Quit martian67 (Read error: 104 (Connection reset by peer))
21:45:42LloreanIt says "the theory is, if we don't frighten users, they might become developers some day"
21:45:48GodEaterno it doesn't
21:46:01GodEaterit says "the more users we have, the more likely developers are to find it"
21:46:08GodEaterby being TOLD about it
21:46:18saratogaGodEater: quit back peddling on the forums
21:46:26LloreanGodEater: You also mentioned them becoming developers.
21:46:30 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
21:46:34GodEateryes some of them might do
21:46:39LloreanGodEater: Either way, though, I don't believe the SVN table immediately makes the website terrifying intrinsically
21:46:40saratogayou were talking about non-technical users then suddenly when I mock you, its about recruiting developers
21:46:40GodEaterbut that's not the thrust of my point
21:46:58GodEatersaratoga: you mocked me? I didn't notice.
21:47:28LloreanGodEater: Your straw poll was between our current page, and the new one, not two versions of the new one, one with a new-user-friendlier SVN log
21:47:42gevaertsWhere do we tell new users about rockbox being GPL?
21:48:03LloreanGodEater: I bet if you showed on with "Recent work" and the SVN log at the bottom most of them wouldn't care much at all which one was presented to them.
21:48:05GodEaterLlorean: I also asked specifically about whether or not they felt the SVN table from the old one should be included on the new one
21:48:08GodEaterthey said no
21:48:19LloreanGodEater: I never said the SVN table from the old one should be on the new one without changes
21:48:32GodEaterwell the svn table IS on the new one with changes
21:48:37GodEaterit's hidden by default
21:48:39GodEaterwhich is great
21:48:47LloreanI disagree
21:48:51GodEaterI can tell :)
21:49:07LloreanAnd I don't think your poll tells anything
21:49:07saratogaat this point I don't even think you have an argument, its jsut a stream of words
21:49:22GodEatersaratoga: if you could read more than two sentences at once I'd care what you thought.
21:49:32LloreanGodEater: You haven't really presented to me any way in which your methodology was unbiased. You showed them a bad page (we all agree it's bad) then you asked if elements of a page they're now inclined to dislike should be added to the good one.
21:49:39*domonoky feels like kindergarden ge
21:49:47saratogai can read lots of words but only if they ever get around to saying something
21:49:49*gevaerts joins domonoky's corner
21:49:59saratogaparagraph after paragraph of nothing
21:50:07JdGordon|FIGHT FIGHT FIGHT!
21:50:16LloreanGodEater: If we're willing to have project news, why *shouldn't* we have "Recent Updates" with just the date/time and commit message?
21:50:50LloreanIt's very much less imposing than an SVN table, and people who don't get technical bits will still skim over
21:51:24GodEaterLlorean: I didn't say it was unbiased. I even said you were free to ignore it.
21:52:38LloreanGodEater: The fact remains that the front page of the site is the first thing many people will see.
21:52:50LloreanA sterilized SVN table is unlikely to be particularly harmful, and can be disguised as news
21:54:10*domonoky thinks a "Recent Changes" (small svn table, only dev, and comment) is a good compromise, and wouldnt hurt anyone..
21:54:16GodEaterI disagree - I think it *is* harmful.
21:54:44*domonoky wonders how such a thing table could cause harm ?
21:54:53Lloreandomonoky: It will scare away people, apparently
21:55:28gevaerts*not* having such a table will probably also scare people away
21:55:33domonokyit will not scare them more then the current website
21:55:42GodEaterwhich will be the bigger number I wonder
21:55:43 Quit martian67 (Remote closed the connection)
21:55:52gevaertsno idea, and no way to know
21:55:57GodEaterindeed not
21:56:10 Quit LambdaCalculus37 (" ajax IRC Client")
21:56:21GodEaterwell - I can see I've lost this argument. But I felt it was worth trying at least.
21:57:57BagderI can see how we need a rock, paper, scissors tournament at devcon to solve this!
21:58:13*GodEater doesn't think anyone would honor the outcome
21:58:18JdGordon|not an exploding frozen coke one? much more fun
21:58:27gevaertsrussian coke?
21:58:53domonokyor maybe this is a job for RSB ? :-)
21:59:08JdGordon|isnt it nearly voting time for the next rsb round?
21:59:08GodEaternot really
21:59:22GodEaterRSB gets involved when there's a tough argument to settle
21:59:27GodEaterno-one's agreeing with me :)
21:59:51JdGordon|GodEater: that brings up another problem here... we always have these arguemtns that dont get an answer... we really need to start just accepting the work by whoever and however it gets done...
22:00:19GodEaterwell I think I have my answer really
22:02:36domonokyso the conclusion is: we want a svn table, but with less info then the current one ? so we can continue on the other website issues .. :-)
22:03:31LloreanGodEater: I do think if it seems like our webpage is still intimidating people if we do put it in there, we should change it.
22:03:40 Quit motionman95 (" ajax IRC Client")
22:03:42GodEaterLlorean: and how will you tell ?
22:04:00LloreanGodEater: Listen to what new users after it approaches think, probably.
22:04:07GodEaterbtw, on a *slightly* related not, has anyone put the google analytics stuff on our page ?
22:04:10LloreanThere will be people who were intimidated but not run off.
22:04:23GodEaterI remain skeptical that that will happen
22:04:30GodEaterwe're bad at listening to users imo
22:05:19LloreanWell, we choose to ignore them a lot
22:05:32LloreanFor example, all of the users who commented in the thread on the issue *like* the SVN table. ;)
22:05:50GodEaterwho by definition are not new...
22:05:58LloreanYes, but they were new, and not run off.
22:06:09GodEaterindeed, but we know there are some that were
22:06:19GodEaterthey're not as visible though
22:06:36LloreanYeah, but we have no real way to know how many were run off.
22:07:07LloreanI personally feel that those that were run off will, with the new sit (with an SVN table at the bottom of the page, sterilized) will, if it's imposing, ignore it and click the download button at the top
22:07:18*GodEater hopes this is true
22:07:54LloreanWhat sort of person takes even the *existence* of technical talk they don't have to read as "too much, gotta leave" exactly?
22:09:04JdGordon|the same that sees a 180page manual and says the same thing
22:09:05GodEaterwho cares?
22:09:11GodEaterthe fact is they exist
22:09:21GodEaterI don't even care why
22:09:41LloreanGodEater: I'm not certain they do in any great quantity. We haven't really established that.
22:09:59GodEaterAnd I'm certain they do.
22:10:26LloreanAnd they want to install Rockbox?
22:10:46GodEaterthe lady who was sitting next to me today expressed great interest in doing so when I told her about it
22:12:36LloreanAnd she is the sort of person who, when presented with mixed content, is unable to skip over a section called "Recent Developments"?
22:13:00GodEaternot unable - it made her believe the product wasn't polished enough to make her want to risk her precious device
22:13:04GodEaterI believe I quoted her :)
22:13:28LloreanIt *is* open source software
22:13:31LloreanIt *is* buggy
22:13:35LloreanIt *is* under active development.
22:13:50*gevaerts wants the site to also say that it *is* GPLv2!
22:13:53GodEaterbut we have releases now
22:14:06LloreanThe releases are still kinda buggy.
22:14:09BigBambiI don't see the problem in having some sort of svn/recent developments whatever at the bottom - that way it is there but not the central eye-catching aspect
22:14:13GodEaterwhich are, imo, no more buggy than an original firmware
22:14:59GodEaterand in many cases a lot less buggy
22:15:16GodEaterI had a comment from a gigabeat F user in my poll who was greatly impressed to learn of Rockbox
22:15:24GodEaterhe'd stopped using his toy he thought the OF was so bad
22:15:31LloreanGodEater: Did she know it might void her warranty? That it might alter the way she could interact with it via iTunes (can't use certain songs / videos, playlisting is wonkey)?
22:16:07*Llorean really wishes GodEater had held off on this poll and done it with a mockup of the new page with a sterilized SVN table.
22:16:28*GodEater can do it again some time when he's posted to a different site :)
22:17:41GodEaterdo we display that info (about warranty voiding etc.) in big letters anywhere else currently ?
22:17:50LloreanBut I mean, it sounds like the "problem" is that an SVN table shows "people are working on it" and some people don't like to see that.
22:17:59GodEaternot that I think it's an issue mind you - all our supported targets are so old now they're likely out of warranty anyway
22:18:43saratogahow is telling me to learn to read even a retort, I quoted you saying those things
22:18:56BigBambicompletely out of context
22:19:01BigBambithey could be from anywhere
22:19:05LloreanGodEater: My point with that was more, even though it's easy, installing and using Rockbox isn't exactly a lowest common denominator action. Presenting it as "just another program" and hiding the fact that it's something constantly being worked on to avoid scaring off people who are afraid of open source development
22:19:32saratogathey're from that thread, and they are in context, and even if they weren't how is a personal attack a reply?
22:19:44GodEaterno it's not a lowest common denominator - I'm not thinking some slack jawed moron is going to find it easy to use
22:19:55BigBambiThey are quoted without any context - there is no way of knowing what they are referring to
22:19:55 Quit lastebil (Read error: 60 (Operation timed out))
22:20:00GodEaterbut there are people who are technology users who can manage that who are still not interested in development information
22:20:16LloreanGodEater: And you think THOSE people would be intimidated by a "Recent Developments" at the bottom of the page?
22:20:17saratogaBigBambi: there is sufficient context
22:20:36BigBambisaratoga: Then we shall have to agree to disagree - I think there is zero context
22:20:42GodEaterif you want to use the word "initimated" then yes
22:21:00saratogathen you are wrong, there is the thread, which provides context
22:21:00LloreanI did use the word intimidated...
22:21:14BigBambiOf the people that would like svn on the page, do you generally agree with me that it would be OK at the bottom?
22:21:19GodEateryou could substitute "put off"
22:21:36BigBambisaratoga: If you have to re-read the thread in order to understand the quotes then the quotes are meaningless
22:21:44LloreanBigBambi: At the bottom, and sterilized (remove the list of source files edited and SVN revision, just show comment and date, and name the table "Recent Development")
22:21:58BigBambiLlorean: Yeah, that's the sort of thing I mean
22:22:00 Join _fml [0] (n=4fd3cac2@gateway/web/cgi-irc/
22:22:45LloreanGodEater: You said she was put off because it made her feel it might be unstable.
22:22:48_fmlHello. Has it been requested before that an FMR file could contain the FM region setting?
22:23:08GodEaterI don't believe she used the word "unstable"
22:23:11LloreanGodEater: I take that as a *good* thing actually. It makes everything less misleading if they realize it's something made by humans, and still being worked on.
22:23:14 Join EternalRains [0] (
22:23:18LloreanGodEater: Unpolished, then
22:23:21LloreanWhich is even better.
22:23:36Llorean_fml: I don't think so, but it sounds like a really good idea to me.
22:23:36GodEaterI think you lack a lot of faith in our work then :(
22:24:02 Quit Thundercloud (Read error: 104 (Connection reset by peer))
22:24:15LloreanGodEater: Someone expecting polish is going to see DivX0 and a hex address and panic. Someone with reasonable expectations about open source software will realize it's just a bug and come talk to us
22:24:37saratogaBigBambi: presumably someone has already read the thread up until then, and remembers him saying those things
22:24:40GodEaterLlorean: are they? I haven't seen a Div 0 error in years
22:24:49LloreanGodEater: I've seen reports of them less than a month ago
22:24:58saratogai mean its not a stretch to assume someone remembers the last 5-10 posts, at least in very reduced form???
22:25:03GodEateryes, but we hardly see them every day
22:25:03LloreanGodEater: Someone thought it was a problem with the "Divx codec"
22:25:25LloreanI still think trying to make the project look like it's *not* under active development for fear of putting people off is misleading
22:25:48saratogain the worst case if they really think i made them up they could do a control + f, though I think most people assume i did not fabricate them
22:25:50BigBambisaratoga: Then why bother quoting them? Without context they could be anything. "non-technical users" could have been in a entance saying they will like it, they will dislike it, we want more, we want less - there is no way to tell from the quotes
22:25:53LloreanIt's made by humans, for humans, and we're working on it constantly. This is a good message to transmit, and such a Recent Developments does it excellently.
22:26:07BigBambisaratoga: You fabricating them isn't the point
22:26:11saratogaBigBambi: to support my point
22:26:17LloreanMuch better than "Last update on..." since that hardly shows the frequency or type of updates.
22:26:28saratogaquoting people is a common way to show what they have previously argued
22:26:45 Join EternalRains1 [0] (
22:26:52saratogaparticularly combined with an explination of why they are being quoted
22:26:54BigBambiOnly if the quote contains onformation
22:26:56saratoga(see my posts)
22:27:04BigBambiWhat you quoted doesn't show which way he argued
22:27:16BigBambiIt only shows that he mentioned non-technical users a few times
22:27:24saratogaI think it does, and if not, why has no one argued otherwise in the thread?
22:27:26 Quit EternalRains (Nick collision from services.)
22:27:34 Nick EternalRains1 is now known as EternalRains (
22:27:36GodEaterLlorean: I don't think anyone visiting the site is going to miss that it's under active development
22:27:37LloreanGodEater: Basically. I see some of your negatives as positives, so yeah, I guess we really can't end up agreeing on this one
22:27:44saratogaif you're asking me to prove that he mentioned them exclusively, thats kind of difficult to prove
22:27:47GodEaterfair enough
22:28:16saratogai can simply prove that numerous mentions exist, kind of hard to prove that nothign else exists without quoting the entire thread
22:28:33LloreanGodEater: There are degrees of activity though. I just think the SVN log can present the most information, in the least frightening way.
22:28:57BigBambisaratoga: It is nothing to do with nothing else existing
22:29:03GodEateras I said in the other channel a "Rockbox last got new features at XX:XX" would do that
22:29:04BigBambisaratoga: Please, read what I said
22:29:09saratogai have
22:29:13GodEaterBigBambi: he can't read
22:29:14GodEatergive up!
22:29:24LloreanGodEater: that doesn't address my slightly earlier statement about frequency or type of activity.
22:29:41GodEaterthe message can be generated off the back of svn commits
22:29:50GodEaterit just doesn't show anything of the kind
22:29:50BigBambisaratoga: Anyway, never mind - lets stop filling the logs with this :)
22:30:03LloreanGodEater: It doesn't show the frequency, nor the location of activity, exactly
22:30:27GodEaterI don't think it has to
22:30:36LloreanIt doesn't _have_ to.
22:30:40GodEaterwe *know* it'll change frequently, because we know how often we get commits
22:30:45LloreanBut it conveys more information to potential developers when it does.
22:31:04GodEaterI feel certain potential developers will find their way to the commit logs
22:31:36LloreanGodEater: We've had people express interest after seeing commits related to players not in our supported players list.
22:31:56LloreanFor example, v2 work going on quite a bit relatively recently
22:32:18*GodEater is missing the point of this bit he thinks
22:32:46LloreanWhere are we lost at?
22:33:10GodEaterwhat people expressing interest in commits related to players not in our supported list has to do with anything :)
22:33:28LloreanThey're the sort of people who may otherwise not see their player in the list, and leave.
22:33:53GodEaterI doubt that too
22:34:03GodEaterwe get enough people enquiring about players not on the front page in the forums
22:34:11LloreanIn my opinion sacrificing a *small* amount of users (and notably - those intimidated by technical things) is better to make things a little more obviously in-development and dev-friendly
22:34:13 Quit moos (Read error: 110 (Connection timed out))
22:34:43LloreanWhat we're talking about is throttling the user word of mouth very slightly to try to look more attractive to people who are actually interested in the development status at a glance.
22:34:56GodEateryes we are
22:35:13LloreanThe great thing about word of mouth is that it *does* spread as long as we're gaining users.
22:35:23LloreanSo we're just discussing the rate of spread, not the existence of it.
22:36:00GodEaterevery new user is a possibly exponential gain in others though
22:36:05GodEaterso it *might* be a big hit
22:36:42LloreanGodEater: Every new user who's intimidated by a Recent Development log, tries Rockbox, finds it's more complex than skipping over reading a table, and tells all his friends it's absolute balls is a potential exponential loss.
22:36:54LloreanSo basically, we're just pulling numbers out of our asses on who it'll gain and lose us.
22:37:04JdGordon|but since when did we care about our userbase?
22:37:07GodEaterhahaha good point
22:37:12GodEaterI can't refute that Llorean :)
22:37:30GodEaterJdGordon|: well apparently we don't - I was arguing we should
22:37:34*Llorean thinks a small barrier of entry to weed out the LCD is better than none.
22:37:40GodEaterit's apparently an unpopular argument
22:37:46LloreanI think we should care about the users, but not at the expense of certain things.
22:38:46LloreanI also think being friendly to non-new users is still a good thing.
22:39:01***Saving seen data "./dancer.seen"
22:39:08*GodEater hopes we'll score some newbies after his survey today
22:39:08LloreanPeople like to drop by the front page and see "yep, the guys are still workin' on it even though there's no news"
22:39:28GodEaterI think we could improve how often we update the news
22:39:32GodEaterwhile we're on the subject
22:39:36planetbeingRockbox seems really developer-oriented. Personally I think that's really awesome, but I guess some people might disagree.
22:39:50 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
22:39:57GodEaterplanetbeing: we're not as bad as your blog for that though ;)
22:40:01planetbeingI'm just wondering how it managed to attract so many developers and keep its irc channel so clean. =P
22:40:08GodEaternot that I object personsally ;)
22:40:21LloreanGodEater: I think the "MajorChanges" new feature category should automatically be news.
22:40:21GodEaterplanetbeing: we kick trouble makers out ruthlessly ;)
22:40:37GodEaterLlorean: yep - good idea
22:40:38planetbeingMy blog is meant to be a little scary. ;)
22:40:48 Join kugel [0] (n=kugel@rockbox/developer/kugel)
22:40:49*GodEater is quaking in his boots
22:41:27kugelthe discussion about the new page is kinda symbolic for how rockbox works, which is sad
22:41:37GodEaterit is ?
22:41:41planetbeingkugel: How so?
22:42:29kugelthere's a proposed feature/new thing, which most if not all people generally agree to
22:42:32saratogawe argue a lot until people lose interest and then something happens
22:43:03BagderI think arguing about matters is a good thing
22:43:13GodEaterminus the name calling probably
22:43:30*GodEater thinks the thread could probably be pruned a bit
22:43:30kugeland then, very few people are prolonging progress, or even make it not happen, by arguing endlessly about small bits of the proposed thing
22:43:32Bagderright, arguing needs to be subject-focused and not people-focused
22:43:44planetbeingDo a lot of new features get indefinitely tabled due to minority dissent?
22:44:01GodEaterplanetbeing: no - we have the Rockbox Steering Board to resolve deadlocks
22:44:27kugelsaratoga: or nothing happens
22:44:28LloreanGodEater: A steering board that so far has never been needed.
22:44:33gevaertsit has never done anything though, so it's uncertain if it works well
22:44:34Bagderreally, the new features we get posted these days are almost all minorities...
22:44:39JdGordon|planetbeing: yes they do
22:44:47LloreanI think that's worth mentioning. When the subject of the RSB comes up, we often start looking for ways to avoid needing to actually ask them
22:45:01GodEaterLlorean: indeed - which is something to be proud of
22:45:05*JdGordon| tends to agree with kugel here
22:45:27Lloreangevaerts: I think it works well *because* it never does anything. At least, as an RSB member there have been a few times where I've tried to work out a compromise I personally didn't agree with to avoid it having to go to it.
22:45:39JdGordon|GodEater: no, i tihnk thats a waste of time... either we have and use the RSB or we argue for way longer than needed and nothing gets done
22:45:40kugelLlorean: I think it was needed once or twice, but people hesitate to ask it, since it would mean admitting that we can't find a solution the normal way
22:46:04GodEaterJdGordon|: well in this case I don't believe that - the www redesign is still moving
22:46:06gevaertsI think we need a formal way for someone to call in the RSB.
22:46:09GodEaterit's just one point we're arguing over
22:46:19linuxstb"RSB: Ping"
22:46:23GodEatergevaerts: let's get a committee together to decide how that's done
22:46:32saratogai think we usually move eventually on most disagreements
22:46:38Lloreangevaerts: Mail the committers list asking for them, I think. *one* person needs to decide it's worth an RSB decision, then propose it, with the RSB being able to say "we think it's not actually ready to come to us yet" or making a choice.
22:46:41 Quit Grahack ("Leaving.")
22:46:43gevaertsGodEater: I'd actually propose to talk about that on devcon :)
22:46:52saratogathe things we havne't moved on are usually poorly implemented patches like mulitfont
22:46:56GodEatergevaerts: excellent
22:47:03*Bagder added RSB as a topic for devcon
22:47:05BigBambisaratoga: true
22:47:11GodEaterI look forward to more petty minded bickering :)
22:47:20BagderGodEater: but with beer!
22:47:25Lloreankugel, JdGordon|: Thinks that are indefinitely tabled are either things that are rejected, or things nobody is willing to fix.
22:47:28saratogaand in a sense deciding to "wait until the patch works well" is a moving on
22:47:37kugelbut things like the numeric/natural sorting support was prolonged endlessly
22:47:37LloreanI don't see how either of those are really things that should go in
22:47:46gevaertskugel: it wasn't
22:47:48Lloreankugel: It wasn't
22:47:52BigBambiBagder: oh dear, argumentative people + beer! :)
22:47:55Bagderkugel: ... and that is a minor feature
22:48:05gevaertsIt was decided that it didn't work well enough to leave enabled for 3.2
22:48:10kugeland really, it still wouldn't be in if I wouldn't have cared about
22:48:23Lloreankugel: And nothing would be in if *someone* didn't care about it
22:48:28LloreanSo that statement's rather silly.
22:48:42JdGordon|Llorean: you know thats not what we are talking about... the RSB is not needed for something which is outright rejected... its needed (and not used) for things which those of us who want the feature and do it are put off by those who wont use the feature and have loud voices
22:48:51saratogathe releasing are becoming annoying like that, by forcing us to cut features from svn like USB and sorting
22:49:11 Quit jgarvey ("Leaving")
22:49:15LloreanJdGordon|: Which thing is put off indefinitely, has a finished, well working patch, and isn't being put in due to loud voices?
22:49:19kugelgevaerts: I was talking about the before it was committed initially
22:49:22Bagdersaratoga: surely they were removed from the branch and not from trunk?
22:49:29LloreanJdGordon|: If there's so many it's a problem, I'm sure you can name quite a few.
22:49:38gevaertssaratoga: both of those were committed just a few weeks before the release
22:49:40JdGordon|i didnt say there were so many...
22:49:50saratogaBagder: yes of course, but its ackward to have features in trunk for a while and then have them not in a release
22:49:53LloreanJdGordon|: If there's only one or two, it's hardly a major problem. And even then, can you name some?
22:50:12Lloreankugel: The fact that it *was* committed kinda disproves the point though.
22:50:13JdGordon|Llorean: the number is irrelevant...we either have and use the RSB, or we dont
22:50:28JdGordon|and yes... 3x3 Vs real touchscreen is the most recent arguemtn
22:50:30LloreanJdGordon|: You haven't attempted to call it on any of these issues though, so...
22:50:30kugelBagder: if it was so minor, why was it so painful to get it in (it was a bit painful for me)
22:50:53Bagderkugel: because it is a minor feature that people have opinions about.
22:51:19BagderI just mean that it being minor means that it's not a big deal that it takes a long time to add
22:51:20kugelLlorean: however, I could've just stop caring about it and it would still rot. and stop caring seems like a common action to me
22:51:47Lloreankugel: So you propose people who don't have a solid interest try to pursue patches they don't care enough about for some reason?
22:52:22JdGordon|and once again we go off on a tangent
22:52:35LloreanJdGordon|: Is 3x3 vs real touchscreen rotting?
22:52:43LloreanI don't really see it as an example of the problem described earlier.
22:52:49gevaertsJdGordon|: I don't think 3x3 Vs real touchscreen is a very good example, because as I see it nobody objects to real touchscreen being available, so it's only a question about what the default setting should be
22:53:02LloreanThere's no real reason they can't work in parallel, and the decision hardly needs to be made now either way.
22:53:22kugel"for some reason" is what I'm complaining about, this reason is that we can't make decisions in a reasonable time frame because of more or less minor complains
22:53:28Lloreangevaerts: And the decision isn't even _due_ yet, so it's hardly being delayed. It doesn't need to be made until we actually support or at least provide builds for a touchscreen target
22:53:49Lloreankugel: You see them as minor. The people making them do not. Your opinion doesn't matter any more or less than theirs.
22:54:44Bagderand someone _could_ call the RSB card at some point
22:54:49Bagderwe could try it...
22:55:01kugelFor me, it seems the new site could die before it lives because of the svn table issue. Not the first time I notice things like this
22:55:16GodEaterkugel: that's utter rubbish
22:55:19GodEaterthe site will go live
22:55:19LloreanBagder: I definitely think the process should just be "Anyone with a patch desiring a decision can call it"
22:55:23 Join SirFunk [0] (
22:55:24BagderI don't think so, as I don't believe Zagor cares ;-)
22:55:26GodEaterit will have an svn table or it won't
22:55:33GodEaterbut it will still go live
22:55:36LloreanGodEater: Exactly.
22:55:41BagderLlorean: I agree
22:55:50gevaertskugel: as far as I'm concerned, the new site could die because of the flexible layout thing, but not because of the svn table...
22:56:17planetbeingHow does the RSB process work? Are there hearings and testimony? =P
22:56:30GodEaterplanetbeing: we don't know - we've never had to use it ;)
22:56:32kugelI'm not saying it dies, but that I fear the the possiblity of it dying. I shouldn't even feel that
22:56:44JdGordon|planetbeing: we'll let you know if they ever meet :p
22:56:46kugelsince we all agreed on the site in general, didn't we?
22:56:46LloreanI imagine the RSB process will be different on a case-by-case basis
22:57:04Lloreankugel: Not really.
22:57:06GodEaterkugel: there are still layout issues to resolve
22:57:11LloreanBut some of us aren't standing in the way despite disagreements.
22:57:19GodEaterindeed not
22:57:21LloreanIt still needs a lot of general polish anyway.
22:57:27GodEaterI'd rather see it go live than not
22:57:27LloreanHave we had blind users test it yet?
22:57:34GodEaterI don't think so
22:57:39GodEaterthat's a BIG test imo
22:57:42gevaertsnot as far as I know. Is it ready enough for that?
22:57:47Lloreangevaerts: Probably not
22:57:51JdGordon|thats just stupid
22:58:00JdGordon|have we had blind tests on the current site?
22:58:06 Quit _fml ("CGI:IRC")
22:58:11BagderI'm sure we have
22:58:19LloreanJdGordon|: We've tried to accommodate problems blind users had with it
22:58:29GodEaterthey find their way to the mailing lists somehow eh Bagder ? :)
22:58:32LloreanAt the very least we should make sure the new site doesn't introduce new ones
22:58:45BagderGodEater: exactly!
22:59:05BigBambididn't someone say that the current one is less than ideal because of lots of tables?
22:59:10Lloreanplanetbeing: The RSB is just five of us we voted into office with vague goals of conflict resolution in the very worst case. I imagine the process will boil down to "okay, here's the patch. Anyone have questions that still need answered? No. Okay, what's everyone's vote." :-P
22:59:16LloreanBigBambi: Yes.
22:59:31LloreanBigBambi: But that was for layout, I don't know about accessibility.
22:59:58LloreanI think, *generally* our current site is considered accessible enough by our blind users.
23:00:04LloreanAt least, every time I've asked about it, I've gotten no responses.
23:00:05gevaertsJdGordon|: it would be silly to go live with something that doesn't work properly due to some easy to fix things
23:00:11GodEaterwe really need someone who is actually blind or sight impaired to test it. We really suck at doing it ourselves.
23:00:11LloreanThen again, I think our blind users all already hate me anyway.
23:00:25LloreanGodEater: It's almost impossible to do it yourself as a sighted person anyway. Habits just get in the way.
23:00:37GodEaterLlorean: didn't our recent mailling list person not find the manual on the current site ?
23:00:44*GodEater seems to think he was blind
23:00:52LloreanGodEater: Yeah, he was blind, and he didn't find the manual
23:00:59LloreanBut that's mostly because he didn't look.
23:01:04LloreanWell, "read"
23:01:04GodEaterhehehe - possibly
23:01:07JdGordon|Llorean: once there is a patch its too late for RSB... really the descision should be made much earlier.. i.e the outcome should be "in principle we will accept/reject this feature/change"... that wastes the persons time much less
23:01:08GodEaterI'm on the fence there :)
23:01:23LloreanGodEater: He was able to find the link when directed to it, so it was clearly accessible enough to find.
23:01:44*GodEater still gives him the benefit of the doubt
23:01:59GodEaterI don't subscribe to that list for a reason
23:01:59LloreanGodEater: Well, all our other blind users can find it. So he may just have a bad browser or screen reader
23:02:00gevaertsthe current proposal handles big minimum fonts badly, but I suspect that the css fixes I posted on the thread fix that
23:02:03GodEateremotions run high there
23:02:38LloreanJdGordon|: It's rather hard to accept a feature without seeing how it works in practice on the player, or without knowing how much performance or RAM cost will go with it in advance.
23:03:42GodEaterLlorean: so the RSB's decision could be "we accept the possiblity of not refusing such a patch out of hand were it to see the light of day perhaps ?" :)
23:03:56LloreanGodEater: Well, I'd say all patches currently in limbo met that condition...
23:04:05GodEatermy point precisely ;)
23:04:18LloreanThat's the problem, a patch can't be accepted fully until it exists. If there's enough info to reject it outright, we don't need the RSB
23:04:39LloreanThe RSB shouldn't be rejecting *or accepting* in advance patches that are gray enough to need RSB for final arbitration.
23:04:52GodEaterJdGordon needs to invent a time machine, so he can bring the patches back from the future, so he can discuss them without having to write them!
23:06:13 Quit bmbl ("Woah!")
23:06:52 Quit BlakeJohnson86 (Remote closed the connection)
23:06:57JdGordon|right, but 1) we care too much about RAM/bin costs, 2) there is no reason why we cant say "in principle, baring technical fubars we will be open to your patch doing this change", 9/10 times the argument is on the feature and not how its implemented
23:07:44LloreanJdGordon|: So what patch is in limbo because we are in argument about the feature rather than the implementation?
23:08:36JdGordon|I'm not saying there *is* one right now... im saying there has been before and will be again....
23:09:14LloreanSo, in other words, decisions were made on all of them without RSB intervention?
23:09:27 Join itcheg [0] (i=62db4767@gateway/web/ajax/
23:09:31LloreanThe RSB's job isn't to make decisions faster. It's to solve the problem when decisions can't be made.
23:09:39JdGordon|no, arguemnts were had and given up on
23:10:24*JdGordon| heading home
23:10:25 Quit JdGordon| (" ajax IRC Client")
23:10:40 Join Seed [0] (
23:10:55 Join spiorf [0] (
23:11:45*Llorean hasn't heard a single example of a patch the RSB should've been involved with yet, though.
23:12:49 Join _Auron|G1_ [0] (
23:13:05*GodEater doesn't believe there's been one
23:13:52LloreanI think there hasn't unless we retask the RSB, and I think that's a bad idea.
23:14:22 Join BlakeJohnson86 [0] (
23:15:22 Quit itcheg (" ajax IRC Client")
23:16:15 Join froggyman [0] (n=47ba40e2@gateway/web/cgi-irc/
23:16:48froggymanwhy are you able to change the clock speed of devices through the debug menu, with rockbox
23:17:00froggymani mean is there a reason for this?
23:17:23LloreanTo test thing
23:17:25LloreanIt's a debug menu.
23:17:59LloreanYou can only pick between the boosted and unboosted speed, and they're both safe.
23:18:06LloreanSo you can't overclock or do anything potentially hardware damaging with it
23:19:48froggymanok, well actually i think there is 3 choices, but yeah they all are <= 80mhz
23:22:12 Quit Mastery ("maybe windows will stop sucking if I restart")
23:28:12 Join itcheg [0] (i=62db4767@gateway/web/ajax/
23:29:54 Join intrados [0] (
23:30:34 Quit froggyman ("CGI:IRC")
23:34:59 Join Sedgewick [0] (
23:35:10CIA-63New commit by bluebrother (r20835): Make sure the global buffers for ipodpatcher and sansapatcher get allocated and freed only once. Fixes segfaults when the bootloader install class was ...
23:36:47 Join JdGordonn [0] (n=irchon@
23:37:44JdGordonnFFs once again stuck on the fwy on the bus not bloody moving :<
23:39:59 Quit JdGordonn (Remote closed the connection)
23:41:29 Quit itcheg (" ajax IRC Client")
23:42:39 Join wincent [0] (
23:44:12wincentAs I did not find this in documentation, I would like to ask: What period does the system tick have?
23:44:51linuxstb100 ticks per second
23:45:14wincentThat was fast :-) Thanks!
23:45:56linuxstb"HZ" is #define'd to be 100 - so if you use it in calculations, use HZ, not 100.
23:46:08 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
23:47:17wincentOf course. I needed to know the period because I will need to figure out how many samples should be calculated by Pure Data core.
23:48:03wincentSo I'll use something like #define SAMPLES_PER_TICK (SAMPLERATE/HZ).
23:49:06 Join itcheg [0] (i=62db4767@gateway/web/ajax/
23:49:58linuxstbHopefully no-one will want 22.050KHz...
23:51:42wincentActually, PdPod runs PureData with this sample rate :-)
23:52:42wincentWorkaround: Calculate one sample more than needed.
23:53:15linuxstbYou should be able to do better than PdPod though ;)
23:54:02wincentIndeed. I think Rockbox has somewhat less overhead than Linux.
23:54:05 Join motionman95 [0] (i=d807e1fe@gateway/web/ajax/
23:55:18motionman95Well, guys, you know how I was having problems with my plugin? Well, I re-wrote it from the ground up, and now it's working! Whoo hoo!
23:58:24 Quit petur (Remote closed the connection)

Previous day | Next day